English(EN) MAEPose: Self-Supervised Spatiotemporal Learning for Human Pose Estimation on mmWave Video

MAEPose 使用自监督学习进行雷达视频人体姿态估计

作者 PulseAugur 编辑部 · [2 个来源] · 2026-04-30 21:23

研究人员开发了 MAEPose，一种使用毫米波雷达视频进行人体姿态估计的新型自监督方法。该方法直接处理频谱图视频，从无标签数据中学习时空表示，与 RGB 方法相比提高了隐私性。MAEPose 表现出显著的性能提升，超越现有基线高达 22.1%，并在有旁观者干扰的情况下仍能保持准确性。 AI

影响引入了一种使用毫米波雷达进行人体姿态估计的隐私保护、自监督方法，可能影响监控和医疗保健应用。

排序理由介绍一种新的人体姿态估计方法的学术论文。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.CV TIER_1 English(EN) · Xijia Wei, Yuan Fang, Kevin Chetty, Youngjun Cho, Nadia Bianchi-Berthouze · 2026-05-04 04:00

MAEPose: Self-Supervised Spatiotemporal Learning for Human Pose Estimation on mmWave Video

arXiv:2605.00242v1 Announce Type: new Abstract: Millimetre-wave (mmWave) radar offers a more privacy-preserving alternative to RGB-based human pose estimation. However, existing methods typically rely on pre-extracted intermediate representations such as sparse point clouds or sp…
arXiv cs.CV TIER_1 English(EN) · Nadia Bianchi-Berthouze · 2026-04-30 21:23

MAEPose: Self-Supervised Spatiotemporal Learning for Human Pose Estimation on mmWave Video

Millimetre-wave (mmWave) radar offers a more privacy-preserving alternative to RGB-based human pose estimation. However, existing methods typically rely on pre-extracted intermediate representations such as sparse point clouds or spectrogram images, where the rich spatiotemporal …